Abstract
Background Dual therapy has emerged as a novel concept for HIV treatment. The ARTE study was aimed at comparing a nucleoside-sparing dual regimen consisting of ATV/r + RAL (DT) vs standard therapy of ATV/r + TDF/FTC (TT) among individuals failing first NNRTI-containing treatment.
Methods Randomized multicenter open-label pilot study. Primary outcome: proportion of subjects with plasma HIV-1 RNA below the limit of detection (<50 copies/mL) at 48 weeks (W48). Secondary outcomes: proportion of discontinuation due to adverse events (AEs), time until viral suppression, time until loss of virological response, development of integrase resistance mutations, and absolute change in CD4 counts. The primary outcome was analyzed using the FDA snapshot analysis.
Results Out of 57 participants screened, 34 were randomized to receive: DT (n: 18) or TT (n: 16). At W48, virological response was achieved in 67% (n: 12/18) of participants receiving DT and 88% (n: 14/16) receiving TT by FDA snapshot analysis (p = NS) and, 73% and 93% by per-protocol analysis (p = NS). CD4 cell count median change from baseline to W48 was +119 and + 52 cell/μL in DT and TT, respectively. Four participants receiving DT and one TT presented virological failure at W48, with low pVL. One participant developed an integrase resistance mutation (155H) and suppressed later on TT.
Conclusion ATV/r+RAL as second-line therapy showed a trend to more frequent virological failure, compared to TT, although the study was unpowered to prove this difference. No major differences were seen in tolerance or toxicity.
This study is registered with ClinicalTrials.gov, Number: NCT01829802
Introduction
Nowadays, the preferred regimen to treat HIV infection, termed combination antiretroviral therapy (cART), consists in combining antiretroviral drugs belonging to, at least, 2 different classes. cART interferes with HIV replication and it is extremely efficient in reducing plasma viral loads to undetectable levels. This results in reduced levels of morbidity and mortality of people with HIV (PWH) and also lower risks of transmission, which in turn impacts the dynamic of the global epidemic1. However, cART cannot cure the infection. Thus, the need for life-long treatment has encouraged the evaluation of simpler regimens containing two drugs (instead of the classical three-drug regimens) to reduce cumulative drug exposure, toxicity, and costs. Indeed, several studies demonstrated the non-inferiority of dual therapy (DT) compared to triple therapy (TT) when used as a switch strategy in virologically suppressed, ART-experienced individuals, or as initial therapy in cART-naive subjects2-5
Here, we aimed at exploring virological response, safety and inflammation markers of a nucleoside-sparing dual regimen consisting of atazanavir/ritonavir (ATV/r) plus raltegravir (RAL) (DT) vs standard therapy of ATV/r + Tenofovir/Emtricitabine (TDF/FTC) (TT) among individuals failing first non-nucleoside reverse transcriptase inhibitors (NNRTI)-containing treatment.
Methods
The ARTE study (Estudio piloto randomizado abierto para evaluar la eficacia y la seguridad de la combinación de Atazanavir/ritonavir + Raltegravir en comparación con Atazanavir/ritonavir más Tenofovir-Emtricitabina en pacientes con infección por VIH que fallaron a un esquema con INNTR) was a pilot, randomized, open-label, two centers, clinical trial (NCT01829802) comparing the effectiveness of an ART regimen of RAL 400mg twice daily + ATV/r 300/100mg once daily to a regimen of TDF/FTC 300/200mg + ATV/r 300/100mg once daily in subjects failing to a NNRTI-based first line ART. ARTE study was approved by the Comité de Bioética de Fundación Huésped FWA IORG0001557, IRB00002014. Participants provided written informed consent before any study procedure. The study was designed to evaluate 50 subjects with a 1:1 ratio distribution amongst both study arms, randomly assigned by a computer-generated randomization table.
Subjects were eligible for enrolment if they were infected with HIV-1, 18 years or older, in good general medical health, not pregnant and willing to use two contraceptive methods, had no abnormal laboratory results and no alcohol or substance misuse, and had no active hepatitis B or C or AIDS-defining conditions.
Subjects also were required to have a documented virological failure (defined as more than 500 copies/ml in two consecutive measures) on a first ART regimen based on 1 NNRTI plus 2 NRTIs, which should have been received without interruption for at least 4 weeks prior to enrolment. Genotyping resistance test at screening should not show resistance to ATV or TDF according to the 2013 IAS-USA guidelines (for ATV mutations I50L, I84V, and N88S, and for TDF mutations K65R, T69S and 2 or more thymidine analogue mutations including M41L o L210W).
The enrollment period was from January to December 2016. Participants were assessed at screening, baseline, and at weeks 4, 8, 12, 24, 36, and 48 and final follow-up visit, or at early termination. Clinical assessments were performed, and laboratory samples were collected at each visit. Mineral Bone density was evaluated at baseline and 48 weeks (W48). Adherence was assessed by pill count and by a self-administered adherence questionnaire at each visit except for the final or follow-up visit.
All laboratory studies were performed at the Fundación Huésped laboratory: viral load was measured utilizing the Abbott Real-time HIV-1 assay (Abbott Molecular, Des Plaines, IL, USA) for plasma HIV-RNA concentrations.
The primary outcome was the proportion of subjects with viral suppression (<50 copies/ml) in an Intention to treat analysis (ITT), at W48. Secondary outcomes were the proportion of subjects discontinuing due to adverse events (AEs), time until viral suppression, time until loss of virological response, the development of integrase resistance mutations, and the absolute change in CD4 count (measured by the difference between basal CD4 count and W48).
Continuous variables were described using medians and interquartile ranges (IQR). Categorical variables were summarized with frequencies and simple proportions. The primary outcomes were analyzed using an FDA snapshot6 and per-protocol analysis. Time until viral suppression and time until loss of virological response were analyzed in the intention-to-treat population using Kaplan-Meyer survival curves and the log-rank test.
Results
Fifty-seven subjects were screened for this study. Twenty-three were excluded (see Figure 1) and the remaining 34 were randomized 1:1 to receive DT (N=18) or TT (N=16). Eighty percent were male and the median age was 38 years old. Baseline characteristics are shown in table 1. No significant differences between groups were identified. At W48, follow-up of two participants were lost, one in each arm. The study had a low recruitment rate, and had to be stopped prematurely due to a higher number of virological failure in the investigational arm, with the additional concern of one case of emergence of integrase resistance mutation. Therefore, we report here, results of the first 34 randomized participants.
Demographic and baseline characteristics of participants per study group
Study flow diagram.
Virological response at W48, by ITT analysis, was 67% (N=12/18) for participants receiving DT and 88% (N= 14/16) for participants receiving TT by FDA snapshot analysis (p=0,233, Figure 2A); and 70% (DT) and 93% (TT) by per-protocol analysis (p=0.099). Three participants (2 in DT and 1 in TT) never reached complete suppression while 2 participants (both in DT) experienced a VL rebound. In total, 5 participants had virological failure at W48, 4 in DT (median VL 390 copies/ml) and 1 in TT (140 copies/ml). One participant from the DT group never achieved virological suppression and developed the INSTI-resistance associated mutation 155H. This participant was discontinued of the study drug and was successfully suppressed with a triple therapy regimen.
Percentage of participants with viral suppression (according to FDA snapshot analysis) (A), time to virological response (B) and time to virological failure (C), per study arm.
There were no significant differences in time to virological response (p=0.238) or in time to virological failure (p=0.279, Figure 2B-C), although the low sample size limits this analysis. CD4 T-cell count median change from baseline to W48 was +119 and +52 cell/□L in DT and TT, respectively (p=0.089).
No deaths were recorded. The main laboratory alteration was hyperbilirubinemia Grade 2 or higher which was seen in 77% of DT and 94% in TT, none requiring stopping ART. Overall, Grade 2 AEs probably related to study drugs were observed in 8/34 (24%) participants: 1 in DT (gastrointestinal symptoms) and 7 in TT (5 gastrointestinal symptoms, 1 renal stone and 1 rash). Finally, 3 Serious AEs (pneumonia and stroke and, Bells paralysis), none related to study drugs, occurred in 2 participants, both in DT.
Regarding the lipid profile and bone mineral density no significant differences were observed between both arms. Of note, bone mineral density showed a lower decrease in DT arm than in TT both when columnar and femur density were evaluated, but the difference was not statistically significant.
Discussion
In times when INSTIs were not widely available, individuals received first-line treatment with NNRTI-based regimen7, and almost all subjects who failed, received a new regimen based on boosted protease inhibitors with two NRTIs. NRTIs continue to be widely used because of their widely proven effectiveness; however, toxicity and resistance can limit its use. Furthermore, NRTI-related toxicity may include lipoatrophy, lipid changes, neuropathy, lactic acidosis, bone and renal toxicity. These side effects are less frequently reported with tenofovir alafenamide.
When this study was designed, there were limited evidence about the safety and effectiveness of dual therapy with ATV/r+RAL8-12. ATV and RAL were an interesting combination to explore in dual therapy due to a potential better metabolic profile of this combination compared to the standard triple therapy at that time based on NNRTI or PIs + 2 NRTIs. These reasons provided the rationale for studying this combination as a second line regimen after failure to a NNRTI based regimen.
Nevertheless, in our study, this dual-therapy regimen presented a higher proportion of virological failure than the triple-therapy arm. In addition, one subject developed a mutation associated with Raltegravir resistance that could compromise the efficacy of second-generation integrase inhibitors (155H), with no emergence of mutations associated to protease inhibitor in the triple-therapy arm.
The rate of side-effects was similar between groups with the exception of gastrointestinal symptoms that were higher in the TT arm. As observed in other tenofovir sparing regimens, there was a lower impact in the decrease of bone mineral density in the DT arm.
The main study limitation includes the limited sample size, as this was designed as a pilot trial, the premature close of the study, and the lack of pharmacokinetic analysis. However, the randomized design of the pilot study, and contemporary published studies revealing similar trends (such as the reports published by Van Lunzen et al13 and Marinaro et al14), support the conclusions.
In conclusion, ATV/r+RAL as second-line therapy, when compared with ATV/r+TDF/FTC, showed a trend towards higher virological failure. Hence, it is unlikely that ATV/r+RAL may provide a significant enough benefit over TT enough to justify its use.
Data Availability
All data produced in the present study are available upon reasonable request to the authors
Disclaimer
The authors participated in the ARTE study, where Fundación Huésped was the sponsor of the study, and MSD provided funds to carry out the study procedures and raltegravir.
Acknowledgments
This study was funded by Merck Co. through investigator-initiated research granted to Fundación Huésped. The authors thank the study participants, study investigators and their teams. We gratefully acknowledge Fabiana Enjamio for data collection, Mariano Destefano for RedCap Database supervicion, and Valeria Alvarez for regulatory and manager support.
References
